{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T20:52:53Z","timestamp":1761598373457},"reference-count":32,"publisher":"Springer Science and Business Media LLC","issue":"6","license":[{"start":{"date-parts":[[2021,9,15]],"date-time":"2021-09-15T00:00:00Z","timestamp":1631664000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2021,9,15]],"date-time":"2021-09-15T00:00:00Z","timestamp":1631664000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"National Key Research and Development Program of China","award":["2017YFB1002504"],"award-info":[{"award-number":["2017YFB1002504"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62073260"],"award-info":[{"award-number":["62073260"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Appl Intell"],"published-print":{"date-parts":[[2022,4]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>In recent years, Deep Neural Networks (DNNs) have achieved excellent performance on many tasks, but it is very difficult to train good models from imbalanced datasets. Creating balanced batches either by majority data down-sampling or by minority data up-sampling can solve the problem in certain cases. However, it may lead to learning process instability and overfitting. In this paper, we propose the Batch Balance Wrapper (BBW), a novel framework which can adapt a general DNN to be well trained from extremely imbalanced datasets with few minority samples. In BBW, two extra network layers are added to the start of a DNN. The layers prevent overfitting of minority samples and improve the expressiveness of the sample distribution of minority samples. Furthermore, Batch Balance (BB), a class-based sampling algorithm, is proposed to make sure the samples in each batch are always balanced during the learning process. We test BBW on three well-known extremely imbalanced datasets with few minority samples. The maximum imbalance ratio reaches 1167:1 with only 16 positive samples. Compared with existing approaches, BBW achieves better classification performance. In addition, BBW-wrapped DNNs are 16.39 times faster, relative to unwrapped DNNs. Moreover, BBW does not require data preprocessing or additional hyper-parameter tuning, operations that may require additional processing time. The experiments prove that BBW can be applied to common applications of extremely imbalanced data with few minority samples, such as the classification of EEG signals, medical images and so on.<\/jats:p>","DOI":"10.1007\/s10489-021-02623-9","type":"journal-article","created":{"date-parts":[[2021,9,15]],"date-time":"2021-09-15T02:37:08Z","timestamp":1631673428000},"page":"6723-6738","update-policy":"http:\/\/dx.doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":9,"title":["BBW: a batch balance wrapper for training deep neural networks on extremely imbalanced datasets with few minority samples"],"prefix":"10.1007","volume":"52","author":[{"given":"Jingzhao","family":"Hu","sequence":"first","affiliation":[]},{"given":"Hao","family":"Zhang","sequence":"additional","affiliation":[]},{"given":"Yang","family":"Liu","sequence":"additional","affiliation":[]},{"given":"Richard","family":"Sutcliffe","sequence":"additional","affiliation":[]},{"given":"Jun","family":"Feng","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2021,9,15]]},"reference":[{"key":"2623_CR1","doi-asserted-by":"crossref","unstructured":"Buda M, Maki A, Mazurowski MA (2017) A systematic study of the class imbalance problem in convolutional neural networks. Neural Networks 106:S0893608018302107\u2013","DOI":"10.1016\/j.neunet.2018.07.011"},{"key":"2623_CR2","doi-asserted-by":"crossref","unstructured":"Mathews L, Hari S (2018) Learning from imbalanced data","DOI":"10.4018\/978-1-5225-2255-3.ch159"},{"issue":"JUN.15","key":"2623_CR3","doi-asserted-by":"publisher","first-page":"137","DOI":"10.1016\/j.knosys.2019.03.001","volume":"174","author":"C Zhang","year":"2019","unstructured":"Zhang C, Bi J, Xu S, Ramentol E, Fan G, Qiao B, Fujita H (2019) Multi-imbalance: An open-source software for multi-class imbalance learning. Knowl.-Based Syst. 174(JUN.15):137\u2013143","journal-title":"Knowl.-Based Syst."},{"key":"2623_CR4","doi-asserted-by":"publisher","unstructured":"Sharma S, Bellinger C, Krawczyk B, Zaiane O, Japkowicz N (2018) Synthetic oversampling with the majority class: A new perspective on handling extreme imbalance. In: 2018 IEEE International Conference on Data Mining (ICDM). https:\/\/doi.org\/10.1109\/ICDM.2018.00060, pp 447\u2013456","DOI":"10.1109\/ICDM.2018.00060"},{"key":"2623_CR5","doi-asserted-by":"crossref","unstructured":"Zheng W, Zhao H (2020) Cost-sensitive hierarchical classification for imbalance classes. Appl Intell 1\u201311","DOI":"10.1007\/s10489-019-01624-z"},{"issue":"1","key":"2623_CR6","doi-asserted-by":"publisher","first-page":"63","DOI":"10.1109\/TKDE.2006.17","volume":"18","author":"ZH Zhou","year":"2006","unstructured":"Zhou Z H, Liu X Y (2006) Training cost-sensitive neural networks with methods addressing the class imbalance problem. IEEE Trans Knowl Data Eng 18(1):63\u201377","journal-title":"IEEE Trans Knowl Data Eng"},{"issue":"12","key":"2623_CR7","doi-asserted-by":"publisher","first-page":"3358","DOI":"10.1016\/j.patcog.2007.04.009","volume":"40","author":"Y Sun","year":"2007","unstructured":"Sun Y, Kamel M S, Wong A K C, Wang Y (2007) Cost-sensitive boosting for classification of imbalanced data. Pattern Recogn. 40(12):3358\u20133378","journal-title":"Pattern Recogn."},{"issue":"Jan.","key":"2623_CR8","first-page":"104837.1","volume":"187","author":"F Zhou","year":"2020","unstructured":"Zhou F, Yang S, Fujita H, Chen D, Wen C (2020) Deep learning fault diagnosis method based on global optimization gan for unbalanced data. Knowle-Based Sys 187(Jan.):104837.1\u2013104837.19","journal-title":"Knowle-Based Sys"},{"issue":"99","key":"2623_CR9","first-page":"2999","volume":"PP","author":"TY Lin","year":"2017","unstructured":"Lin T Y, Goyal P, Girshick R, He K, Dollar P (2017) Focal loss for dense object detection. IEEE Trans Pattern Anal Mach Intell PP(99):2999\u20133007","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2623_CR10","doi-asserted-by":"crossref","unstructured":"Li B, Liu Y, Wang X (2018) Gradient harmonized single-stage detector. arXiv preprint arXiv:181105181","DOI":"10.1609\/aaai.v33i01.33018577"},{"key":"2623_CR11","doi-asserted-by":"publisher","first-page":"349","DOI":"10.1016\/j.ymssp.2018.03.025","volume":"110","author":"F Jia","year":"2018","unstructured":"Jia F, Lei Y, Lu N, Xing S (2018) Deep normalized convolutional neural network for imbalanced fault classification of machinery and its understanding via visualization. Mech. Syst. Signal Process. 110:349\u2013367","journal-title":"Mech. Syst. Signal Process."},{"issue":"99","key":"2623_CR12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/JSTARS.2020.3022997","volume":"PP","author":"L Zhang","year":"2020","unstructured":"Zhang L, Zhang C, Xiao H, Quan S, Liu L (2020) A class imbalance loss for imbalanced object recognition. IEEE J Sel Top Appl Earth Obs Remote Sens PP(99):1\u20131","journal-title":"IEEE J Sel Top Appl Earth Obs Remote Sens"},{"key":"2623_CR13","doi-asserted-by":"publisher","first-page":"660","DOI":"10.1016\/j.procs.2020.09.038","volume":"176","author":"I Valova","year":"2020","unstructured":"Valova I, Harris C, Mai T, Gueorguieva N (2020) Optimization of convolutional neural networks for imbalanced set classification. Procedia Computer Science 176:660\u2013669","journal-title":"Procedia Computer Science"},{"key":"2623_CR14","unstructured":"Zhang C, Kjellstrom H, Mandt S (2017) Determinantal point processes for mini-batch diversification. arXiv preprint arXiv:170500607"},{"key":"2623_CR15","unstructured":"Qi Q, Xu Y, Jin R, Yin W, Yang T (2020) Attentional biased stochastic gradient for imbalanced classification. arXiv preprint arXiv:201206951"},{"key":"2623_CR16","unstructured":"Shoeb AH (2009) Application of machine learning to epileptic seizure onset detection and treatment. Massachusetts Institute of Technology"},{"issue":"6","key":"2623_CR17","doi-asserted-by":"publisher","first-page":"061907","DOI":"10.1103\/PhysRevE.64.061907","volume":"64","author":"RG Andrzejak","year":"2001","unstructured":"Andrzejak R G, Lehnertz K, Mormann F, Rieke C, David P, Elger C E (2001) Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state. Phys Rev E Stat Nonlin Soft Matter Phys 64(6):061907","journal-title":"Phys Rev E Stat Nonlin Soft Matter Phys"},{"key":"2623_CR18","doi-asserted-by":"crossref","unstructured":"Xie Y, Wu Z, Han X, Wang H, Wu Y, Cui L, Feng J, Zhu Z, Chen Z (2020) Computer-aided system for the detection of multicategory pulmonary tuberculosis in radiographs. Journal of Healthcare Engineering","DOI":"10.1155\/2020\/9205082"},{"key":"2623_CR19","doi-asserted-by":"crossref","unstructured":"Huang G, Liu Z, Pleiss G, Van Der Maaten L, Weinberger K (2019) Convolutional networks with dense connectivity. IEEE Trans Pattern Anal Mach Intell","DOI":"10.1109\/TPAMI.2019.2918284"},{"issue":"04","key":"2623_CR20","doi-asserted-by":"publisher","first-page":"687","DOI":"10.1142\/S0218001409007326","volume":"23","author":"Y Sun","year":"2011","unstructured":"Sun Y, Wong A K C, Kamel M S (2011) Classification of imbalanced data: A review. Int J Pattern Recognit Artif Intell 23(04):687\u2013719","journal-title":"Int J Pattern Recognit Artif Intell"},{"issue":"9","key":"2623_CR21","doi-asserted-by":"publisher","first-page":"1263","DOI":"10.1109\/TKDE.2008.239","volume":"21","author":"H He","year":"2009","unstructured":"He H, Garcia E A (2009) Learning from imbalanced data. IEEE Trans Knowl Data Eng 21 (9):1263\u20131284","journal-title":"IEEE Trans Knowl Data Eng"},{"issue":"1","key":"2623_CR22","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1613\/jair.953","volume":"16","author":"NV Chawla","year":"2002","unstructured":"Chawla N V, Bowyer K W, Hall L O, Kegelmeyer W P (2002) Smote: Synthetic minority over-sampling technique. J. Artif. Intell. Res. 16(1):321\u2013357","journal-title":"J. Artif. Intell. Res."},{"key":"2623_CR23","doi-asserted-by":"crossref","unstructured":"Han H, Wang W Y, Mao B H (2005) Borderline-smote: A new over-sampling method in imbalanced data sets learning. In: International conference on advances in intelligent computing","DOI":"10.1007\/11538059_91"},{"key":"2623_CR24","doi-asserted-by":"crossref","unstructured":"Bunkhumpornpat C, Sinapiromsaran K, Lursinsap C (2009) Safe-level-smote: Safe-level-synthetic minority over-sampling technique for handling the class imbalanced problem. In: Pacific-asia conference on advances in knowledge discovery & data mining","DOI":"10.1007\/978-3-642-01307-2_43"},{"key":"2623_CR25","doi-asserted-by":"publisher","first-page":"128","DOI":"10.1016\/j.inffus.2019.07.006","volume":"54","author":"J Sun","year":"2020","unstructured":"Sun J, Li H, Fujita H, Fu B, Ai W (2020) Class-imbalanced dynamic financial distress prediction based on adaboost-svm ensemble combined with smote and time weighting. Information Fusion 54:128\u2013144","journal-title":"Information Fusion"},{"issue":"2","key":"2623_CR26","doi-asserted-by":"publisher","first-page":"539","DOI":"10.1109\/TSMCB.2008.2007853","volume":"39","author":"L Xu-Ying","year":"2009","unstructured":"Xu-Ying L, Jianxin W, Zhi-Hua Z (2009) Exploratory undersampling for class-imbalance learning. IEEE Trans Sys Man & Cybern Part B 39(2):539\u2013550","journal-title":"IEEE Trans Sys Man & Cybern Part B"},{"issue":"8","key":"2623_CR27","doi-asserted-by":"publisher","first-page":"2807","DOI":"10.1007\/s10489-019-01423-6","volume":"49","author":"P Lopez-Garcia","year":"2019","unstructured":"Lopez-Garcia P, Masegosa A D, Osaba E, Onieva E, Perallos A (2019) Ensemble classification for imbalanced data based on feature space partitioning and hybrid metaheuristics. Appl. Intell. 49 (8):2807\u20132822","journal-title":"Appl. Intell."},{"key":"2623_CR28","doi-asserted-by":"crossref","unstructured":"Hayashi T, Ambai K, Fujita H (2020) Applying cluster-based zero-shot classifier to data imbalance problems","DOI":"10.1007\/s10489-021-02671-1"},{"key":"2623_CR29","doi-asserted-by":"publisher","first-page":"106034","DOI":"10.1109\/ACCESS.2019.2931865","volume":"7","author":"JS Lee","year":"2019","unstructured":"Lee JS (2019) Auc4.5: Auc-based c4.5 decision tree algorithm for imbalanced data classification. IEEE Access 7:106034\u2013106042","journal-title":"IEEE Access"},{"key":"2623_CR30","doi-asserted-by":"publisher","first-page":"351","DOI":"10.1016\/j.neucom.2020.03.064","volume":"404","author":"A Taherkhani","year":"2020","unstructured":"Taherkhani A, Cosma G, McGinnity T (2020) Adaboost-cnn: An adaptive boosting algorithm for convolutional neural networks to classify multi-class imbalanced datasets using transfer learning. Neurocomputing 404:351\u2013366. https:\/\/doi.org\/10.1016\/j.neucom.2020.03.064","journal-title":"Neurocomputing"},{"key":"2623_CR31","doi-asserted-by":"crossref","unstructured":"P\u00e9rez-Hern\u00e1ndez F, Tabik S, Lamas A, Olmos R, Herrera F (2020) Object detection binary classifiers methodology based on deep learning to identify small objects handled similarly: Application in video surveillance. Knowledge-Based Systems, p 105590","DOI":"10.1016\/j.knosys.2020.105590"},{"key":"2623_CR32","unstructured":"Ioffe S, Szegedy C (2015) Batch normalization: Accelerating deep network training by reducing internal covariate shift, pp 448\u2013456"}],"container-title":["Applied Intelligence"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10489-021-02623-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10489-021-02623-9\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10489-021-02623-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,4,13]],"date-time":"2022-04-13T04:26:01Z","timestamp":1649823961000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10489-021-02623-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9,15]]},"references-count":32,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2022,4]]}},"alternative-id":["2623"],"URL":"https:\/\/doi.org\/10.1007\/s10489-021-02623-9","relation":{},"ISSN":["0924-669X","1573-7497"],"issn-type":[{"value":"0924-669X","type":"print"},{"value":"1573-7497","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,9,15]]},"assertion":[{"value":"15 June 2021","order":1,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"15 September 2021","order":2,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that they have no conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"<!--Emphasis Type='Bold' removed-->Conflict of Interests"}}]}}